We have studied the molecular basis for virulence in hepatitis A virus. Virulence (and attenuation) is controlled principally by two genes: VP1/2A and 2C. However, attenuating mutations are strongly selected against in vivo, resulting in the emergence of virulent variants. This has important implications for the development of a live attenuated hepatitis A vaccine. The genetic heterogeneity of hepatitis C virus is believed to play an important role in its pathogenicity. To examine this relationship, we have determined the genetic heterogeneity of HCV isolates that were recovered from patients who were infected following transfusion. The sequence of the hypervariable region and adjacent portions of envelope proteins 1 and 2 were determined for multiple clones obtained from patients who had fulminant hepatitis, from patients who convalesced following acute hepatitis and from patients who progressed to chronic hepatitis C. Distinctive patterns of dynamic change in the sequence of clones during the first several weeks of infection were observed. Patients with fulminant or resolving hepatitis had few changes in the sequences of clones, whereas there were many changes in the sequences of clones from patients who progressed to chronic hepatitis. Thus, the outcome of an HCV infection could be predicted in the first few weeks of the infection. Similar studies have been carried out in patients chronically infected with HCV and who were undergoing therapy with interferon. Patients could be separated into four groups, based upon their response to interferon therapy: a) long-term responders, b) those who relapsed following cessation of treatment, c) those who responded but had a break-through of viral replication while still on therapy and d) those who failed to respond (nonresponders). As with acutely infected HCV patients, distinctive patterns of dynamic change in the sequence and heterogeneity of HCV clones obtained early in therapy were predictive of outcome. Long-term responders demonstrated a marked decrease in heterogeneity of HCV, resulting in eradication of the virus. Relapsing patients also demonstrated a change in heterogeneity and a decrease in viral titer but a new dominant strain usually emerged following cessation of therapy. Nonresponders maintained the same dominant strain throughout therapy, suggesting that an interferon-resistant strain already existed before therapy. Patients who experienced a breakthrough during therapy had patterns that were similar to those of nonresponders, suggesting that this was a mixed group of patients. These findings may be useful in predicting the outcome of therapy with interferon early in the course of treatment. More recently, we have extended these studies to an analysis HCV sequences in infants and young children who were infected at birth. Although such perinatal transmissions are rare, they do offer an opportunity to study the interaction of the developing host immune response in an immunologically naive infant. Transmission of HCV from mother to infant often is characterized by virtually monoclonal viral infection initially, probably resulting from immunological selection by the mother's immune system. As the infant mounts its own immunological response, the genomic sequence becomes heterogeneous. Thus, the progression from monoclonal to polyclonal viral populations resembles that seen in chimpanzees experimentally infected with monoclonal HCV. Although considerable information has been gained from these longitudinal studies of patients, it is difficult to study the mechanisms of pathogenesis in such systems. Chimpanzees, which are the only animals other than man that are susceptible to infection with HCV provide an experimental model for studying the interactions of the host and the virus in the pathogenesis of hepatitis. Collaborative studies with Frank Chisari, Scripps Institute, have demonstrated that, both in hepatitis B virus infections and hepatitis C virus infections, the cellular immune response plays an important role in noncytolytic down-regulation of viral replication and cytolytic removal of residual infected cells. These two mechanisms are sequential and overlapping and the former appears to be mediated by interferon gamma and the latter by CD8+ CTLs and, perhaps, by interferon gamma through its proinflammatory activity. These studies have also revealed that type 1 interferon (interferon alpha/beta)-activated antiviral proteins are expressed in response to the viral infections, but that the viruses, especially HCV, are resistant to the antiviral activity of this innate immune response. Microarray studies of the host's immune responses to viral hepatitis and how the hepatitis viruses attempt to circumvent those responses are currently in progress. Among the human hepatitis viruses, only certain hepatitis E virus strains can replicate in nonprimate species of animals. We have studied the comparative pathogenesis of a swine-derived strain and a human-derived strain in swine as well as in nonhuman primates. These studies are important because of the potential for zoonotic spread of hepatitis E virus.